A Framework Using Contrastive Learning for Classification with Noisy Labels

Madalina Ciortan, Romain Dupuis and Thomas Peel
Additional contact information
Madalina Ciortan: R&D Department, EURA NOVA, 1435 Mont-Saint-Guibert, Belgium
Romain Dupuis: R&D Department, EURA NOVA, 1435 Mont-Saint-Guibert, Belgium
Thomas Peel: R&D Department, EURA NOVA, 1435 Mont-Saint-Guibert, Belgium

Data, 2021, vol. 6, issue 6, 1-26

Abstract: We propose a framework using contrastive learning as a pre-training task to perform image classification in the presence of noisy labels. Recent strategies, such as pseudo-labeling, sample selection with Gaussian Mixture models, and weighted supervised contrastive learning have, been combined into a fine-tuning phase following the pre-training. In this paper, we provide an extensive empirical study showing that a preliminary contrastive learning step brings a significant gain in performance when using different loss functions: non robust, robust, and early-learning regularized. Our experiments performed on standard benchmarks and real-world datasets demonstrate that: (i) the contrastive pre-training increases the robustness of any loss function to noisy labels and (ii) the additional fine-tuning phase can further improve accuracy, but at the cost of additional complexity.

Keywords: noisy labels; image classification; contrastive learning; robust loss (search for similar items in EconPapers)
JEL-codes: C8 C80 C81 C82 C83 (search for similar items in EconPapers)
Date: 2021
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2306-5729/6/6/61/pdf (application/pdf)
https://www.mdpi.com/2306-5729/6/6/61/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jdataj:v:6:y:2021:i:6:p:61-:d:571848

Access Statistics for this article

Data is currently edited by Ms. Cecilia Yang

More articles in Data from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().